Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I was finally able to get a trace from a 'task hung for 120 seconds' error. Our servers would just randomly crash on load spikes. IO wait times seem normal. Memory/Swap is fine. These are Xen based Ubuntu 10.04 - 64bit. VPS, some are even "Baremetals".

Below is the trace:

Aug 24 11:13:25 admin1 kernel: [2732107.501749] INFO: task java:29952 blocked for more than 120 seconds.
Aug 24 11:13:25 admin1 kernel: [2732107.501954] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Aug 24 11:13:25 admin1 kernel: [2732107.502223] java          D 0000000000000000     0 29952      1 0x00000000
Aug 24 11:13:25 admin1 kernel: [2732107.502227]  ffff8801d3b5fbb8 0000000000000082 ffffffff81a35800 ffff8801d635be8c
Aug 24 11:13:25 admin1 kernel: [2732107.502230]  0000000000014cc0 ffff88014fcc5b80 0000000000014cc0 ffff88014fcc5f18
Aug 24 11:13:25 admin1 kernel: [2732107.502233]  ffff8801d3b5ffd8 ffff88014fcc5f20 ffff8801d3b5e010 0000000000014cc0
Aug 24 11:13:25 admin1 kernel: [2732107.502236] Call Trace:
Aug 24 11:13:25 admin1 kernel: [2732107.502243]  [<ffffffff81589bd7>] __mutex_lock_slowpath+0xf7/0x180
Aug 24 11:13:25 admin1 kernel: [2732107.502247]  [<ffffffff81589abb>] mutex_lock+0x2b/0x50
Aug 24 11:13:25 admin1 kernel: [2732107.502251]  [<ffffffff8122633a>] __log_wait_for_space+0xda/0x1b0
Aug 24 11:13:25 admin1 kernel: [2732107.502255]  [<ffffffff814fed2f>] ? inet_recvmsg+0x4f/0x80
Aug 24 11:13:25 admin1 kernel: [2732107.502258]  [<ffffffff812220e1>] start_this_handle+0xf1/0x440
Aug 24 11:13:25 admin1 kernel: [2732107.502260]  [<ffffffff812225c1>] ? journal_start+0x81/0x100
Aug 24 11:13:25 admin1 kernel: [2732107.502263]  [<ffffffff812225f5>] journal_start+0xb5/0x100
Aug 24 11:13:25 admin1 kernel: [2732107.502266]  [<ffffffff811d3e71>] ext3_journal_start_sb+0x31/0x60
Aug 24 11:13:25 admin1 kernel: [2732107.502269]  [<ffffffff811ca63d>] ext3_dirty_inode+0x3d/0xa0
Aug 24 11:13:25 admin1 kernel: [2732107.502273]  [<ffffffff811776ca>] __mark_inode_dirty+0x3a/0x1f0
Aug 24 11:13:25 admin1 kernel: [2732107.502275]  [<ffffffff811ca7f5>] ext3_setattr+0x155/0x2c0
Aug 24 11:13:25 admin1 kernel: [2732107.502279]  [<ffffffff810097ce>] ? __switch_to+0x26e/0x320
Aug 24 11:13:25 admin1 kernel: [2732107.502283]  [<ffffffff8116c2c5>] notify_change+0x175/0x2f0
Aug 24 11:13:25 admin1 kernel: [2732107.502285]  [<ffffffff81588781>] ? schedule+0x411/0xa50
Aug 24 11:13:25 admin1 kernel: [2732107.502289]  [<ffffffff81153064>] do_truncate+0x64/0xa0
Aug 24 11:13:25 admin1 kernel: [2732107.502291]  [<ffffffff81153d7b>] ? generic_file_llseek+0x5b/0x70
Aug 24 11:13:25 admin1 kernel: [2732107.502294]  [<ffffffff811532d8>] sys_ftruncate+0xe8/0xf0
Aug 24 11:13:25 admin1 kernel: [2732107.502297]  [<ffffffff8100b0b2>] system_call_fastpath+0x16/0x1b

and another just below now when 'unzip' was run - exactly the same time as the above trace;

Aug 24 11:13:25 admin1 kernel: [2732107.503946] unzip         D 0000000000000000     0 28123      1 0x00000000
Aug 24 11:13:25 admin1 kernel: [2732107.503951]  ffff880031e156e8 0000000000000086 0000000000000000 0000000000000000
Aug 24 11:13:25 admin1 kernel: [2732107.503954]  0000000000014cc0 ffff8801183f5b80 0000000000014cc0 ffff8801183f5f18
Aug 24 11:13:25 admin1 kernel: [2732107.503957]  ffff880031e15fd8 ffff8801183f5f20 ffff880031e14010 0000000000014cc0
Aug 24 11:13:25 admin1 kernel: [2732107.503960] Call Trace:
Aug 24 11:13:25 admin1 kernel: [2732107.503964]  [<ffffffff81227745>] log_wait_commit+0xc5/0x150
Aug 24 11:13:25 admin1 kernel: [2732107.503967]  [<ffffffff81081150>] ? autoremove_wake_function+0x0/0x40
Aug 24 11:13:25 admin1 kernel: [2732107.503969]  [<ffffffff8158b08e>] ? _raw_spin_lock+0xe/0x20
Aug 24 11:13:25 admin1 kernel: [2732107.503972]  [<ffffffff81225ddb>] log_do_checkpoint+0xab/0x530
Aug 24 11:13:25 admin1 kernel: [2732107.503976]  [<ffffffff81036bd6>] ? __ticket_spin_lock+0x16/0x20
Aug 24 11:13:25 admin1 kernel: [2732107.503979]  [<ffffffff81226306>] __log_wait_for_space+0xa6/0x1b0
Aug 24 11:13:25 admin1 kernel: [2732107.503982]  [<ffffffff81070974>] ? mod_timer+0x174/0x2d0
Aug 24 11:13:25 admin1 kernel: [2732107.503985]  [<ffffffff812220e1>] start_this_handle+0xf1/0x440
Aug 24 11:13:25 admin1 kernel: [2732107.503987]  [<ffffffff812225c1>] ? journal_start+0x81/0x100
Aug 24 11:13:25 admin1 kernel: [2732107.503990]  [<ffffffff812225f5>] journal_start+0xb5/0x100
Aug 24 11:13:25 admin1 kernel: [2732107.503993]  [<ffffffff811d3e71>] ext3_journal_start_sb+0x31/0x60
Aug 24 11:13:25 admin1 kernel: [2732107.503995]  [<ffffffff811ccbe2>] ext3_write_begin+0xa2/0x230
Aug 24 11:13:25 admin1 kernel: [2732107.503998]  [<ffffffff8110029e>] generic_file_buffered_write+0xfe/0x250
Aug 24 11:13:25 admin1 kernel: [2732107.504001]  [<ffffffff811dc65f>] ? ext3_xattr_security_get+0x2f/0x40
Aug 24 11:13:25 admin1 kernel: [2732107.504004]  [<ffffffff81102df0>] __generic_file_aio_write+0x240/0x470
Aug 24 11:13:25 admin1 kernel: [2732107.504006]  [<ffffffff81103085>] generic_file_aio_write+0x65/0xd0
Aug 24 11:13:25 admin1 kernel: [2732107.504009]  [<ffffffff8115370a>] do_sync_write+0xda/0x120
Aug 24 11:13:25 admin1 kernel: [2732107.504012]  [<ffffffff812941a8>] ? apparmor_file_permission+0x18/0x20
Aug 24 11:13:25 admin1 kernel: [2732107.504015]  [<ffffffff8126366f>] ? security_file_permission+0x1f/0x80
Aug 24 11:13:25 admin1 kernel: [2732107.504017]  [<ffffffff811539f8>] vfs_write+0xc8/0x190
Aug 24 11:13:25 admin1 kernel: [2732107.504020]  [<ffffffff81154251>] sys_write+0x51/0x90
Aug 24 11:13:25 admin1 kernel: [2732107.504023]  [<ffffffff8100b0b2>] system_call_fastpath+0x16/0x1b

can anyone help?

Is this a deadlock on the filesystem?

tia.

share|improve this question
add comment

1 Answer 1

Those stack traces both indicate that a filesystem write operation was being attempted at the time, so check storage related issues. If you have network mounted filesystems check that the server is accessible and performing normally. If not, check your disk drives. You should see messages in dmesg in this case, and use smartmontools to check the state of the drives.

share|improve this answer
    
I have 10 servers which started via ltsp and using the same nfs server but there is no the same issue on the another servers. –  alterpub Dec 11 '12 at 10:49
    
I have checked dmesg on the basePart server and I didn't find any records for current date. –  alterpub Dec 11 '12 at 10:58
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.